Frontier models are failing one in three production attempts — and getting harder to audit

Stanford's 2026 AI Index: frontier models fail one in three attempts, lab transparency is declining, and benchmarks are ...
OfficeChai

China’s Z.AI Releases GLM-5.1, Beats All US Models On SWE-Bench Pro

A Chinese model is now best in the world at a crucial coding benchmark. Z.AI, the Beijing-based lab formerly known as Zhipu ...
OfficeChai

Anthropic’s Claude Mythos Preview Smashes Coding Benchmarks, Scores 77.8 On SWE-Bench Pro

Anthropic is maintaining its lead in coding models, and how. Claude Mythos Preview — the unreleased frontier model at the center of ...
VentureBeat

Arthur unveils Bench, an open-source AI model evaluator

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More New York City-based artificial intelligence (AI) startup Arthur has ...
Live Science

Scientists design new 'AGI benchmark' that indicates whether any future AI model could cause 'catastrophic harm'

It's quick and easy to access Live Science Plus, simply enter your email below. We'll send you a confirmation and sign you up for our daily newsletter, keeping you up to date with the latest science ...
Unite.AI

MiniMax Open Sources M2.7, a Self-Evolving Agent Model

Chinese AI company MiniMax has released the weights for MiniMax M2.7, a 229-billion-parameter Mixture-of-Experts model that participated in its own development cycle – marking what the company calls ...
Morning Overview on MSN

OpenAI launches GPT-Rosalind, a biology-focused model for lab workflows

OpenAI has released GPT-Rosalind, a large language model fine-tuned specifically for life sciences research, marking the ...